- 16 Feb, 2023 40 commits
-
-
Paolo Abeni authored
Pedro Tammela says: ==================== net/sched: transition actions to pcpu stats and rcu Following the work done for act_pedit[0], transition the remaining tc actions to percpu stats and rcu, whenever possible. Percpu stats make updating the action stats very cheap, while combining it with rcu action parameters makes it possible to get rid of the per action lock in the datapath. For act_connmark and act_nat we run the following tests: - tc filter add dev ens2f0 ingress matchall action connmark - tc filter add dev ens2f0 ingress matchall action nat ingress any 10.10.10.10 Our setup consists of a 26 cores Intel CPU and a 25G NIC. We use TRex to shoot 10mpps TCP packets and take perf measurements. Both actions improved performance as expected since the datapath lock disappeared. For act_pedit we move the drop counter to percpu, when available. For act_gate we move the counters to percpu, when available. [0] https://lore.kernel.org/all/20230131145149.3776656-1-pctammela@mojatatu.com/ ==================== Link: https://lore.kernel.org/r/20230214211534.735718-1-pctammela@mojatatu.comSigned-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Pedro Tammela authored
Since act_pedit now has access to percpu counters, use the tcf_action_inc_overlimit_qstats wrapper that will use the percpu counter whenever they are available. Reviewed-by:
Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
Pedro Tammela <pctammela@mojatatu.com> Signed-off-by:
-
Pedro Tammela authored
The tc action act_gate was using shared stats, move it to percpu stats. tdc results: 1..12 ok 1 5153 - Add gate action with priority and sched-entry ok 2 7189 - Add gate action with base-time ok 3 a721 - Add gate action with cycle-time ok 4 c029 - Add gate action with cycle-time-ext ok 5 3719 - Replace gate base-time action ok 6 d821 - Delete gate action with valid index ok 7 3128 - Delete gate action with invalid index ok 8 7837 - List gate actions ok 9 9273 - Flush gate actions ok 10 c829 - Add gate action with duplicate index ok 11 3043 - Add gate action with invalid index ok 12 2930 - Add gate action with cookie Reviewed-by:
-
Pedro Tammela authored
The tc action act_connmark was using shared stats and taking the per action lock in the datapath. Improve it by using percpu stats and rcu. perf before: - 13.55% tcf_connmark_act - 81.18% _raw_spin_lock 80.46% native_queued_spin_lock_slowpath perf after: - 2.85% tcf_connmark_act tdc results: 1..15 ok 1 2002 - Add valid connmark action with defaults ok 2 56a5 - Add valid connmark action with control pass ok 3 7c66 - Add valid connmark action with control drop ok 4 a913 - Add valid connmark action with control pipe ok 5 bdd8 - Add valid connmark action with control reclassify ok 6 b8be - Add valid connmark action with control continue ok 7 d8a6 - Add valid connmark action with control jump ok 8 aae8 - Add valid connmark action with zone argument ok 9 2f0b - Add valid connmark action with invalid zone argument ok 10 9305 - Add connmark action with unsupported argument ok 11 71ca - Add valid connmark action and replace it ok 12 5f8f - Add valid connmark action with cookie ok 13 c506 - Replace connmark with invalid goto chain control ok 14 6571 - Delete connmark action with valid index ok 15 3426 - Delete connmark action with invalid index Reviewed-by: -
Pedro Tammela authored
The tc action act_nat was using shared stats and taking the per action lock in the datapath. Improve it by using percpu stats and rcu. perf before: - 10.48% tcf_nat_act - 81.83% _raw_spin_lock 81.08% native_queued_spin_lock_slowpath perf after: - 0.48% tcf_nat_act tdc results: 1..27 ok 1 7565 - Add nat action on ingress with default control action ok 2 fd79 - Add nat action on ingress with pipe control action ok 3 eab9 - Add nat action on ingress with continue control action ok 4 c53a - Add nat action on ingress with reclassify control action ok 5 76c9 - Add nat action on ingress with jump control action ok 6 24c6 - Add nat action on ingress with drop control action ok 7 2120 - Add nat action on ingress with maximum index value ok 8 3e9d - Add nat action on ingress with invalid index value ok 9 f6c9 - Add nat action on ingress with invalid IP address ok 10 be25 - Add nat action on ingress with invalid argument ok 11 a7bd - Add nat action on ingress with DEFAULT IP address ok 12 ee1e - Add nat action on ingress with ANY IP address ok 13 1de8 - Add nat action on ingress with ALL IP address ok 14 8dba - Add nat action on egress with default control action ok 15 19a7 - Add nat action on egress with pipe control action ok 16 f1d9 - Add nat action on egress with continue control action ok 17 6d4a - Add nat action on egress with reclassify control action ok 18 b313 - Add nat action on egress with jump control action ok 19 d9fc - Add nat action on egress with drop control action ok 20 a895 - Add nat action on egress with DEFAULT IP address ok 21 2572 - Add nat action on egress with ANY IP address ok 22 37f3 - Add nat action on egress with ALL IP address ok 23 6054 - Add nat action on egress with cookie ok 24 79d6 - Add nat action on ingress with cookie ok 25 4b12 - Replace nat action with invalid goto chain control ok 26 b811 - Delete nat action with valid index ok 27 a521 - Delete nat action with invalid index Reviewed-by: -
Paolo Abeni authored
Jesse Brandeburg says: ==================== net/core: commmon prints for promisc Add a print to the kernel log for allmulticast entry and exit, and standardize the print for entry and exit of promiscuous mode. These prints are useful to both user and developer and should have the triggering driver/bus/device info that netdev_info (optionally) gives. ==================== Link: https://lore.kernel.org/r/20230214210117.23123-1-jesse.brandeburg@intel.comSigned-off-by:
-
Jesse Brandeburg authored
The kernel stack can be more consistent by printing the IFF_PROMISC aka promiscuous enable/disable messages with the standard netdev_info message which can include bus and driver info as well as the device. typical command usage from user space looks like: ip link set eth0 promisc <on|off> But lots of utilities such as bridge, tcpdump, etc put the interface into promiscuous mode. old message: [ 406.034418] device eth0 entered promiscuous mode [ 408.424703] device eth0 left promiscuous mode new message: [ 406.034431] ice 0000:17:00.0 eth0: entered promiscuous mode [ 408.424715] ice 0000:17:00.0 eth0: left promiscuous mode Signed-off-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by:
-
Jesse Brandeburg authored
When the user sets or clears the IFF_ALLMULTI flag in the netdev, there are no log messages printed to the kernel log to indicate anything happened. This is inexplicably different from most other dev->flags changes, and could suprise the user. Typically this occurs from user-space when a user: ip link set eth0 allmulticast <on|off> However, other devices like bridge set allmulticast as well, and many other flows might trigger entry into allmulticast as well. The new message uses the standard netdev_info print and looks like: [ 413.246110] ixgbe 0000:17:00.0 eth0: entered allmulticast mode [ 415.977184] ixgbe 0000:17:00.0 eth0: left allmulticast mode Signed-off-by:
-
Paolo Abeni authored
Jamal Hadi Salim says: ==================== net/sched: Retire some tc qdiscs and classifiers The CBQ + dsmark qdiscs and the tcindex + rsvp classifiers have served us for over 2 decades. Unfortunately, they have not been getting much attention due to reduced usage. While we dont have a good metric for tabulating how much use a specific kernel feature gets, for these specific features we observed that some of the functionality has been broken for some time and no users complained. In addition, syzkaller has been going to town on most of these and finding issues; and while we have been fixing those issues, at times it becomes obvious that we would need to perform bigger surgeries to resolve things found while getting a syzkaller fix in place. After some discussion we feel that in order to reduce the maintenance burden it is best to retire them. This patchset leaves the UAPI alone. I could send another version which deletes the UAPI as well. AFAIK, this has not been done before - so it wasnt clear what how to handle UAPI. It seems legit to just delete it but we would need to coordinate with iproute2 (given they sync up with kernel uapi headers). There are probably other users we don't know of that copy kernel headers. If folks feel differently I will resend the patches deleting UAPI for these qdiscs and classifiers. I will start another thread on iproute2 before sending any patches to iproute2. ==================== Link: https://lore.kernel.org/r/20230214134915.199004-1-jhs@mojatatu.comSigned-off-by:
-
Jamal Hadi Salim authored
The rsvp classifier has served us well for about a quarter of a century but has has not been getting much maintenance attention due to lack of known users. Signed-off-by:
Jiri Pirko <jiri@nvidia.com> Signed-off-by:
-
Jamal Hadi Salim authored
The tcindex classifier has served us well for about a quarter of a century but has not been getting much TLC due to lack of known users. Most recently it has become easy prey to syzkaller. For this reason, we are retiring it. Signed-off-by:
-
Jamal Hadi Salim authored
The dsmark qdisc has served us well over the years for diffserv but has not been getting much attention due to other more popular approaches to do diffserv services. Most recently it has become a shooting target for syzkaller. For this reason, we are retiring it. Signed-off-by:
-
Jamal Hadi Salim authored
The ATM qdisc has served us well over the years but has not been getting much TLC due to lack of known users. Most recently it has become a shooting target for syzkaller. For this reason, we are retiring it. Signed-off-by:
-
Jamal Hadi Salim authored
While this amazing qdisc has served us well over the years it has not been getting any tender love and care and has bitrotted over time. It has become mostly a shooting target for syzkaller lately. For this reason, we are retiring it. Goodbye CBQ - we loved you. Signed-off-by:
-
Paolo Abeni authored
Steen Hegelund says: ==================== Adding Sparx5 ES0 VCAP support This provides the Egress Stage 0 (ES0) VCAP (Versatile Content-Aware Processor) support for the Sparx5 platform. The ES0 VCAP is an Egress Access Control VCAP that uses frame keyfields and previously classified keyfields to add, rewrite or remove VLAN tags on the egress frames, and is therefore often referred to as the rewriter. The ES0 VCAP also supports trapping frames to the host. The ES0 VCAP has 1 lookup accessible with this chain id: - chain 10000000: ES0 Lookup 0 The ES0 VCAP does not do traffic classification to select a keyset, but it does have two keysets that can be used on all traffic. For now only the ISDX keyset is used. The ES0 VCAP can match on an ISDX key (Ingress Service Index) as one of the frame metadata keyfields, similar to the ES2 VCAP. The ES0 VCAP uses external counters in the XQS (statistics) group. ==================== Link: https://lore.kernel.org/r/20230214104049.1553059-1-steen.hegelund@microchip.comSigned-off-by:
-
Steen Hegelund authored
This provides these 3 actions for rule in the ES0 VCAP: - action vlan pop - action vlan modify id X priority Y - action vlan push id X priority Y protocol Z Signed-off-by:
Steen Hegelund <steen.hegelund@microchip.com> Signed-off-by:
-
Steen Hegelund authored
This enables the TC command to use the Sparx5 ES0 VCAP, and handling of rule links between IS0 and ES0. Signed-off-by:
-
Steen Hegelund authored
This adds the ES0 VCAP port keyset configuration for Sparx5. Signed-off-by:
-
Steen Hegelund authored
This provides access to the ES0 VCAP register targets Signed-off-by:
-
Steen Hegelund authored
This provides the VCAP model for the Sparx5 ES0 (Egress Stage 0) VCAP. This VCAP provides rewriting functionality in the egress path. Signed-off-by:
-
Steen Hegelund authored
Ensure that an error is returned if the VCAP instance was not found. The chain offset (diff) is allowed to be zero as this just means that the user did not request rules to be linked. Signed-off-by:
-
Steen Hegelund authored
This improves the check performed on linked rules when enabling or disabling them. The chain id used must be the chain id without the offset used for linking the rules. Signed-off-by:
-
Steen Hegelund authored
This changes the TPID of the egress frames to use the TPID stored in the IFH (internal frame header), which ensures that this is the TPID classified for the frame at ingress. Signed-off-by:
-
Steen Hegelund authored
The rule counter must be cleared when creating a new rule, even if the VCAP lookup is currently disabled. This ensures that rules located in VCAPs that use external counters (such as Sparx5 IS2 and ES0) will have their counter reset even if the VCAP lookup is not enabled at the moment. Signed-off-by:
-
Steen Hegelund authored
A valid frame should never use a multicast address as its source MAC address, so discard these invalid frames. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queueJakub Kicinski authored
Tony Nguyen says: ==================== Intel Wired LAN Driver Updates 2023-02-14 (ice) This series contains updates to ice driver only. Karol extends support for GPIO pins to E823 devices. Daniel Vacek stops processing of PTP packets when link is down. Pawel adds support for BIG TCP for IPv6. Tony changes return type of ice_vsi_realloc_stat_arrays() as it always returns success. Zhu Yanjun updates kdoc stating supported TLVs. * '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue: ice: Mention CEE DCBX in code comment ice: Change ice_vsi_realloc_stat_arrays() to void ice: add support BIG TCP on IPv6 ice/ptp: fix the PTP worker retrying indefinitely if the link went down ice: Add GPIO pin support for E823 products ==================== Link: https://lore.kernel.org/r/20230214213003.2117125-1-anthony.l.nguyen@intel.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Randy Dunlap authored
Correct spelling problems for Documentation/core-api/packing.rst as reported by codespell. Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Cc: Vladimir Oltean <olteanv@gmail.com> Cc: Jonathan Corbet <corbet@lwn.net> Reviewed-by:
Mukesh Ojha <quic_mojha@quicinc.com> Acked-by:
Daniel Jordan <daniel.m.jordan@oracle.com> Link: https://lore.kernel.org/r/20230215053738.11562-1-rdunlap@infradead.orgSigned-off-by:
-
Andrew Halaney authored
The schema had snps,reset-delay-us as dependent on snps,reset-gpio. The actual property is called snps,reset-delays-us, so fix this to catch any devicetree defining snsps,reset-delays-us without snps,reset-gpio. Fixes: 7db3545a ("dt-bindings: net: stmmac: Convert the binding to a schemas") Reviewed-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by:
Andrew Halaney <ahalaney@redhat.com> Link: https://lore.kernel.org/r/20230214171505.224602-1-ahalaney@redhat.comSigned-off-by:
-
Davide Caratti authored
remove temporary files created by 'mirred_egress_to_ingress_tcp' test in the cleanup() handler. Also, change variable names to avoid clashing with globals from lib.sh. Suggested-by:
Davide Caratti <dcaratti@redhat.com> Link: https://lore.kernel.org/r/091649045a017fc00095ecbb75884e5681f7025f.1676368027.git.dcaratti@redhat.comSigned-off-by:
-
Mengyuan Lou authored
Add the basic ethtool ops get_drvinfo and get_link for ngbe and txgbe. Ngbe implements get_link_ksettings, nway_reset and set_link_ksettings for free using phylib code. The code related to the physical interface is not yet fully implemented in txgbe using phylink code. So do not implement get_link_ksettings, nway_reset and set_link_ksettings in txgbe. Signed-off-by:
Mengyuan Lou <mengyuanlou@net-swift.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/20230214091527.69943-1-mengyuanlou@net-swift.comSigned-off-by:
-
Willem de Bruijn authored
MSG_ZEROCOPY ensures that pinned user pages do not exceed the limit. If no limit is set, skip this accounting as otherwise expensive atomic_long operations are called for no reason. This accounting is already skipped for privileged (CAP_IPC_LOCK) users. Rely on the same mechanism: if no mmp->user is set, mm_unaccount_pinned_pages does not decrement either. Tested by running tools/testing/selftests/net/msg_zerocopy.sh with an unprivileged user for the TXMODE binary: ip netns exec "${NS1}" sudo -u "{$USER}" "${BIN}" "-${IP}" ... Signed-off-by:Willem de Bruijn <willemb@google.com> Reviewed-by:
Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20230214155740.3448763-1-willemdebruijn.kernel@gmail.comSigned-off-by:
-
Oleksij Rempel authored
Fix warning: drivers/net/phy/phy-c45.c:712 genphy_c45_write_eee_adv() error: uninitialized symbol 'changed' Reported-by:
kernel test robot <lkp@intel.com> Reported-by:
Dan Carpenter <error27@gmail.com> Link: https://lore.kernel.org/r/202302150232.q6idsV8s-lkp@intel.com/ Fixes: 022c3f87 ("net: phy: add genphy_c45_ethtool_get/set_eee() support") Signed-off-by:
Oleksij Rempel <o.rempel@pengutronix.de> Reviewed-by:
-
Dan Carpenter authored
These booleans are never set to false, but are just used without being initialized. Fixes: 4ac94f72 ("net: phy: Add driver for Motorcomm yt8531 gigabit ethernet phy") Signed-off-by:
Frank Sae <Frank.Sae@motor-comm.com> Link: https://lore.kernel.org/r/Y+xd2yJet2ImHLoQ@kiliSigned-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linuxJakub Kicinski authored
Saeed Mahameed says: ==================== mlx5-updates-2023-02-10 1) From Roi and Mark: MultiPort eswitch support MultiPort E-Switch builds on newer hardware's capabilities and introduces a mode where a single E-Switch is used and all the vports and physical ports on the NIC are connected to it. The new mode will allow in the future a decrease in the memory used by the driver and advanced features that aren't possible today. This represents a big change in the current E-Switch implantation in mlx5. Currently, by default, each E-Switch manager manages its E-Switch. Steering rules in each E-Switch can only forward traffic to the native physical port associated with that E-Switch. While there are ways to target non-native physical ports, for example using a bond or via special TC rules. None of the ways allows a user to configure the driver to operate by default in such a mode nor can the driver decide to move to this mode by default as it's user configuration-driven right now. While MultiPort E-Switch single FDB mode is the preferred mode, older generations of ConnectX hardware couldn't support this mode so it was never implemented. Now that there is capable hardware present, start the transition to having this mode by default. Introduce a devlink parameter to control MultiPort Eswitch single FDB mode. This will allow users to select this mode on their system right now and in the future will allow the driver to move to this mode by default. 2) From Jiri: Improvements and fixes for mlx5 netdev's devlink logic 2.1) Cleanups related to mlx5's devlink port logic 2.2) Move devlink port registration to be done before netdev alloc 2.3) Create auxdev devlink instance in the same ns as parent devlink 2.4) Suspend auxiliary devices only in case of PCI device suspend * tag 'mlx5-updates-2023-02-10' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux: net/mlx5: Suspend auxiliary devices only in case of PCI device suspend net/mlx5: Remove "recovery" arg from mlx5_load_one() function net/mlx5e: Create auxdev devlink instance in the same ns as parent devlink net/mlx5e: Move devlink port registration to be done before netdev alloc net/mlx5e: Move dl_port to struct mlx5e_dev net/mlx5e: Replace usage of mlx5e_devlink_get_dl_port() by netdev->devlink_port net/mlx5e: Pass mdev to mlx5e_devlink_port_register() net/mlx5: Remove outdated comment net/mlx5e: TC, Remove redundant parse_attr argument net/mlx5e: Use a simpler comparison for uplink rep net/mlx5: Lag, Add single RDMA device in multiport mode net/mlx5: Lag, set different uplink vport metadata in multiport eswitch mode net/mlx5: E-Switch, rename bond update function to be reused net/mlx5e: TC, Add peer flow in mpesw mode net/mlx5: Lag, Control MultiPort E-Switch single FDB mode ==================== Link: https://lore.kernel.org/r/20230214221239.159033-1-saeed@kernel.orgSigned-off-by:
-
Jakub Kicinski authored
Moshe Shemesh says: ==================== devlink: cleanups and move devlink health functionality to separate file This patchset moves devlink health callbacks, helpers and related code from leftover.c to new file health.c. About 1.3K LoC are moved by this patchset, covering all devlink health functionality. In addition this patchset includes a couple of small cleanups in devlink health code and documentation update. ==================== Link: https://lore.kernel.org/r/1676392686-405892-1-git-send-email-moshe@nvidia.comSigned-off-by:
-
Moshe Shemesh authored
Fix a bug in trace point definition for devlink health report, as TP_STRUCT_entry of reporter_name should get reporter_name and not msg. Note no fixes tag as this is a harmless bug as both reporter_name and msg are strings and TP_fast_assign for this entry is correct. Signed-off-by:
Moshe Shemesh <moshe@nvidia.com> Reviewed-by:
-
Moshe Shemesh authored
Update devlink-health.rst file: - Add devlink formatted message (fmsg) API documentation. - Add auto-dump as a condition to do dump once error reported. - Expand OOB to clarify this acronym. Signed-off-by:
-
Moshe Shemesh authored
Now that all devlink health callbacks and related code are in file health.c move common health functions and devlink_health_reporter struct to be local in health.c file. Signed-off-by:
-
Moshe Shemesh authored
Move devlink health report test callback from leftover.c to health.c. No functional change in this patch. Signed-off-by:
-
Moshe Shemesh authored
Move devlink health report dump callbacks and related code from leftover.c to health.c. No functional change in this patch. Signed-off-by:
-
